CN116739645A - Order abnormity supervision system based on enterprise management - Google Patents

Order abnormity supervision system based on enterprise management Download PDF

Info

Publication number
CN116739645A
CN116739645A CN202310921273.9A CN202310921273A CN116739645A CN 116739645 A CN116739645 A CN 116739645A CN 202310921273 A CN202310921273 A CN 202310921273A CN 116739645 A CN116739645 A CN 116739645A
Authority
CN
China
Prior art keywords
feature space
clustering
order
data
region
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310921273.9A
Other languages
Chinese (zh)
Inventor
韩道峰
姜凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zibo Seagrass Software Service Co ltd
Original Assignee
Zibo Seagrass Software Service Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zibo Seagrass Software Service Co ltd filed Critical Zibo Seagrass Software Service Co ltd
Priority to CN202310921273.9A priority Critical patent/CN116739645A/en
Publication of CN116739645A publication Critical patent/CN116739645A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
    • G06Q30/0635—Processing of requisition or of purchase orders
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Game Theory and Decision Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention relates to the technical field of data processing, in particular to an order abnormity supervision system based on enterprise management, which comprises the following components: acquiring order information data of enterprises; analyzing each index data in the order information data of the enterprise to construct a clustering feature space; performing self-adaptive division on the clustering feature space, setting an optimal clustering parameter rule, and obtaining optimal clustering parameters; and clustering the order information data of the enterprise according to the optimal clustering parameters to obtain abnormal data. The invention improves the uncertainty of the clustering effect of the DBSCAN algorithm in the enterprise order anomaly monitoring scene; analyzing and selecting parameters by combining scenes, and obtaining optimal clustering parameters in a self-adaptive manner; the clustering effect of the DBSCAN clustering algorithm in the current order anomaly monitoring scene is guaranteed; the purpose of accurately monitoring order abnormality is achieved. And optimizing a supervision system.

Description

Order abnormity supervision system based on enterprise management
Technical Field
The invention relates to the technical field of data processing, in particular to an order abnormity supervision system based on enterprise management.
Background
Order anomalies are a concern in enterprise management; which may lead to problems such as insufficient inventory, funds problems and customer satisfaction drops; the abnormality of the order can cause the stock level to fluctuate, which brings difficulty to stock management and increases the risk of backlog stock and backlog; it may also lead to enterprises facing direct financial losses; abnormal orders even involve illegal activities such as fraud, which is likely to lead to enterprises facing legal disputes, regulatory penalties and compliance risks. In order to reduce the harm and influence caused by order abnormality, enterprises generally establish internal examination limit, and manual examination or other modes are adopted to detect and early warn abnormal conditions in the prior art; and corresponding measures are quickly taken. At present, the traditional monitoring mode of the enterprises on the order abnormality mainly adopts a mode of manual examination and threshold setting statistical analysis, and the manual examination is generally only applicable to small-scale enterprises, but the monitoring efficiency and accuracy of the method cannot be ensured along with the increase of the number of the orders; a second common way is to set a threshold for the business for certain key indicators, when a certain indicator of an order exceeds the threshold; the system can automatically trigger early warning, but the method can not recognize and hide complex abnormal conditions therein, namely, can not develop analysis and judgment aiming at specific abnormalities. The threshold setting method can only perform abnormality monitoring on one type of indexes, and does not link and analyze the indexes, so that the real abnormality degree of each abnormal data cannot be obtained.
Therefore, when more enterprises adopt artificial intelligence, algorithms in machine learning monitor and manage the abnormality of order data; wherein data clustering algorithms are often usedAbnormal data is screened, which is a density-based clustering algorithm, and the defects of the traditional monitoring method can be effectively overcome; the construction of the feature space and the selection of the parameter points in the algorithm are difficult to select and evaluate, and different parameter settings can bring about clustering effects with great difference due to the strong sensitivity of the algorithm to the parameters, so that the screening result of abnormal data is inaccurate.
Disclosure of Invention
The invention provides an order abnormity supervision system based on enterprise management, which aims to solve the existing problems.
The order abnormity supervision system based on enterprise management adopts the following technical scheme:
one embodiment of the present invention provides an order anomaly supervision system based on enterprise management, the system comprising:
and a data acquisition module: acquiring order information data of enterprises;
and (3) constructing a feature space module: analyzing each index data in the order information data of the enterprise to construct a clustering feature space;
the optimal parameter obtaining module is used for obtaining: obtaining a dividing node value according to each index data in the order information data; the method comprises the steps of adaptively obtaining a partition area according to a partition node numerical value pair clustering feature space structure, and obtaining a plurality of clustering feature space areas and serial numbers of each clustering feature space area; selecting different adjacent data points according to the sequence numbers of the clustering feature space regions, and obtaining the neighborhood radius in the optimal clustering parameters of any clustering feature space region according to the distance between the adjacent data points; obtaining the minimum data point number in the neighborhood radius region in the optimal clustering parameter of any clustering feature space region according to the sequence number of the clustering feature space region;
and a data detection module: and clustering the order information data of the enterprise according to the neighborhood radius in the optimal clustering parameters of the clustering feature space region and the minimum data point number in the neighborhood radius region to obtain abnormal data, thereby realizing abnormal supervision and management of the order of the enterprise.
Preferably, the step of analyzing each index data in the order information data of the enterprise to construct a cluster feature space includes the following specific steps:
normalizing the index data return rate to serve as a vertical axis in the clustering feature space; taking the ratio of the total amount of the index data order to the number of the order as a horizontal axis in the clustering feature space; the vertical axis and the horizontal axis form a cluster feature space.
Preferably, the obtaining the dividing node value according to each index data in the order information data includes the following specific formulas:
in the method, in the process of the invention,respectively representing the values of dividing nodes along the horizontal and vertical axes in the clustering feature space, < >>Representing the number of all order numbers of order information data of enterprises; />The +.f in order information data representing the enterprise>Characteristic index data return rate corresponding to the order numbers; />The +.f in order information data representing the enterprise>The number of the characteristic index data orders corresponding to the order numbers; />The +.f in order information data representing the enterprise>The characteristic index data corresponding to the order numbers are the order transaction amount; />Is a linear normalization function.
Preferably, the method for obtaining the plurality of cluster feature space regions and the sequence number of each cluster feature space region is as follows:
setting upSequence number of spatial region for clustering feature +.>And (3) taking the value: />The method comprises the steps of carrying out a first treatment on the surface of the Set->The time represents the upper right region in the clustering feature space; set->Representing a region at the lower right part in the clustering feature space; set->The upper left region of the cluster feature space is represented; set->Representing a region at the left lower part in the clustering feature space; thereby obtaining the serial numbers of the four clustering feature space areas and the clustering feature space areas.
Preferably, the selecting different adjacent data points according to the serial numbers of the clustering feature space regions, and obtaining the neighborhood radius in the optimal clustering parameters of any clustering feature space region according to the distance between the adjacent data points includes the following specific steps:
presetting a parameterFor any region in the cluster feature space +.>Acquiring the clustering feature space region +.>Inner every->Data points>To its nearest->The Euclidean distance of the nearest neighbor data point, denoted as +.>Distance sets of data points, wherein Euclidean distances in the distance sets of all data points are sequenced in ascending order to form a set +.>The method comprises the steps of carrying out a first treatment on the surface of the Will collect->The element in the cluster is subjected to difference between adjacent element values from left to right, the maximum value of the difference result is obtained, and the Euclidean distance corresponding to the maximum value is used as the clustering feature space region +.>Is>。
Preferably, the obtaining the minimum number of data points in the neighborhood radius region in the optimal clustering parameter of any clustering feature space region according to the sequence number of the clustering feature space region includes the following specific steps:
the first in the cluster feature spaceThe calculation expression of the minimum data point number in the neighborhood radius area in each area is as follows:
in the method, in the process of the invention,the +.>The minimum number of data points in the neighborhood radius area in each area; />Representing the +.>A plurality of regions; />Representing the dimension of the cluster feature space.
The technical scheme of the invention has the beneficial effects that: the invention is forThe algorithm improves the uncertainty of the clustering effect in the enterprise order anomaly monitoring scene; analyzing and selecting parameters by combining scenes, and obtaining optimal clustering parameters in a self-adaptive manner; ensure->Clustering effect of the clustering algorithm under the current order anomaly monitoring scene; the purpose of accurately monitoring order abnormality is achieved. And optimizing a supervision system.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of an enterprise management based order anomaly supervision system according to the present invention;
FIG. 2 is a schematic diagram of the clustering feature space partitioning results of the enterprise management-based order anomaly monitoring system.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the order anomaly monitoring system based on enterprise management according to the invention with reference to the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the order anomaly monitoring system based on enterprise management provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of an order anomaly supervision system based on enterprise management according to one embodiment of the present invention is shown, where the system includes the following blocks:
and a data acquisition module: order information data of an enterprise is obtained.
It should be noted that, when the order information form stored in the enterprise system is acquired; wherein important index information data need to be contained; such as: order number, customerIndex data such as order amount, order quantity, return rate and the like, and index information data corresponding to each order number and clients. If a large abnormality exists in the numerical value of a certain index corresponding to the order number of any customer, the order of the customer is considered to be an abnormal order, and when different index data corresponding to the order number of the customer is abnormal, the influence degree of the abnormality is also different.
Specifically, each index data corresponding to the order number in the order information data of the enterprise is stored according to the time sequence of the order number; i.e. set upThe collection is used for storing the acquired order amount data sequence,, />the +.f in order information data representing the enterprise>The order transaction amount corresponding to the individual order number,representing the total number of order numbers in order information data of an enterprise; build->The aggregate is used for storing the acquired order quantity data sequence +.>，/>The +.f in order information data representing the enterprise>Order number corresponding to the order number, +.>Representing the total number of order numbers in order information data of an enterprise; thereby collecting and storing the order information data of the enterprise.
Thus, the acquisition of the order information data of the enterprise is completed.
And (3) constructing a feature space module: and analyzing each index data in the order information data of the enterprise to construct a clustering feature space.
It should be noted that, the construction of the front-stage feature space of the clustering algorithm has a great influence on the whole clustering result, and the clustering effect obtained by constructing the horizontal and vertical coordinates in the feature space is difficult to ensure by only selecting any two related indexes in the traditional mode, so that the embodiment needs to analyze the influence degree of the key index data of each order abnormality to comprehensively obtain the feature space construction parameters and adaptively obtain the optimal clustering parameters for the feature space, so that the abnormality with smaller influence degree of the key index data of the order abnormality can be classified as a non-outlier, and the abnormality with larger influence degree of the key index data of the order abnormality can be classified as an outlier; and obtaining more accurate classification results, thereby realizing personalized and effective abnormal supervision and management.
The construction of the cluster feature space generally requires the determination of a comparison parameter, i.e., the determination of a horizontal axis and vertical axis index; a generally more efficient way is to select two characteristic index data related to an order anomaly; for example, the horizontal axis is set as the order amount, and the vertical axis is set as the order quantity; in order to combine the anomaly monitoring scene of the implementation and obtain the optimal clustering effect under the scene, each index data in the order information data of the enterprise needs to be analyzed.
The specific analysis is as follows: if there is a sudden increase or decrease in the amount of orders in an order anomaly, which is an abnormal fluctuation in the amount of orders, this problem is typically not severe due to sales promotion, seasonal demand or supply chain problems, which are more normal market changes or customer purchase differences. For abnormal order amount in the index, namely, the order amount obviously deviates from the normal range, the abnormal order amount is probably caused by incorrect price quotation, incorrect discount calculation or malicious tampering, so that the abnormal order is in a serious condition; for the abnormal return rate data in the index, the abnormal return rate data is also a serious order abnormal phenomenon, and the abnormal return rate data is a sudden rise of the return rate of a certain product, which implies the quality problem, descriptive disagreement or physical distribution damage of the product. Therefore, the embodiment needs to link the three types of characteristic index data; and comprehensively evaluating order abnormity.
Taking the three characteristic index data as an example, the three characteristic index data are expressed in a two-dimensional characteristic space by using a dimension reduction idea, so that the abnormal attribute of the order information data of an enterprise can be objectively and effectively reflected; thereby obtaining the clustering feature space parameters.
Specifically, the feature index data return rate is normalized and then used as a vertical axis representation in the clustering feature space, namelyThe method comprises the steps of carrying out a first treatment on the surface of the The ratio of the total amount of the orders to the number of the orders of the characteristic index data is taken as the horizontal axis in the clustering characteristic space to represent +.>。
The saidRepresenting the vertical axis in the cluster feature space; />Representing the horizontal axis in the cluster feature space, +.>If the product return rate corresponding to any order is higher, the larger the marked data point vertical axis data of the order in the feature space is, the more the marked data point vertical axis data is positioned above the feature space, and otherwise, the marked data point vertical axis data is positioned below the feature space; />Representing the total amount of the characteristic index data order; />Representing the number of orders of the characteristic index data; />Then is the ratio of the two>Is a linear normalization function; mapping the result to a value range { 0-1 }; the larger value indicates a higher individual amount of the order, when the larger the value of the horizontal axis of the marked data point of the order in the feature space, i.e. to the right of the feature space; if the ratio is smaller, the individual amounts representing the order are smaller, i.e., the marked data points for the order in the feature space are to the left of the feature space.
It should be further noted that, the method for constructing the clustering feature space in this embodiment has better effect than the traditional method for constructing the clustering feature space, and can analyze the specific condition of abnormal indexes of the data points through the distribution condition of the data points in the feature space; and then, the self-adaptive clustering rule parameter adjustment is carried out on the different indexes according to the abnormal influence degrees, so that a better personalized clustering effect is achieved, and more serious order abnormalities are accurately identified.
Thus, the construction of the clustering feature space is completed.
The optimal parameter obtaining module is used for obtaining: and carrying out self-adaptive division on the clustering feature space, setting an optimal clustering parameter rule, and obtaining optimal clustering parameters.
1. And carrying out self-adaptive division on the clustering feature space.
It should be noted that, because the abnormal order information data occupies a smaller ratio in the whole order information data; taking the average value of the horizontal axis index data values and the average value of the vertical axis index data values of all orders as a division standard; the more the order index data corresponding to the nodes deviating from the dividing value is more likely to be abnormal data points; thereby dividing the cluster feature space into regions.
Specifically, the partitioned areas are obtained in a self-adaptive mode according to the clustering feature space construction principle; obtaining a partition value node threshold value:
in the method, in the process of the invention,respectively representing the values of dividing nodes along the horizontal and vertical axes in the clustering feature space, < >>Representing the number of all order numbers of order information data of enterprises; />The +.f in order information data representing the enterprise>Characteristic index data return rate corresponding to the order numbers; />The +.f in order information data representing the enterprise>The number of the characteristic index data orders corresponding to the order numbers; />The +.f in order information data representing the enterprise>The characteristic index data corresponding to the order numbers are the order transaction amount; />Is a linear normalization function.
Dividing the clustering feature space according to the obtained clustering feature space dividing nodes, wherein a clustering feature space dividing result schematic diagram is shown in fig. 2; in the figureDividing nodes for clustering feature space>And (5) the sequence number of the cluster feature space region.
Thus, the self-adaptive division of the clustering feature space is completed, and four areas of the clustering feature space and the sequence numbers of the areas of the clustering feature space are obtained.
2. And obtaining the neighborhood radius in the optimal clustering parameters of any region of the clustering feature space.
The clustering feature space is subjected to regional division to the extent that the clustering feature space is affected byRepresenting partitioned cluster feature spatial regions, different cluster feature spatial regions corresponding to +.>The values of the cluster feature space region are different, and the influence degrees of the cluster feature space region are different.
Setting upFor divided areas>And (3) taking the value: />The method comprises the steps of carrying out a first treatment on the surface of the Different values of the three-dimensional space represent different space regions and are setThe upper right area in the clustering feature space is represented, the area is an area with higher return rate and higher order amount, and abnormal data points in the area have great influence on the order information data of enterprises; set->Representing a region at the lower right part in the clustering feature space; the area is higher in order amount, and affects less than the abnormal data points in the upper right area; setting upThe upper left region of the cluster feature space is represented; the area is higher in return rate but relatively normal in order amount, the area abnormal data point effect is relatively normal, and +.>Representing a region at the left lower part in the clustering feature space; the area is an area with relatively normal order amount and low return rate, and abnormal data points of the area have little influence on the order information data of enterprises.
Presetting a parameterWherein the present embodiment is +.>Examples are described, the present embodiment is not particularly limited, wherein +.>Depending on the particular implementation.
Specifically, setting optimal clustering parameters for each region in a clustering feature space; the rule of the set neighborhood radius is as follows:
for any region in the cluster feature spaceAcquiring the clustering feature space region +.>Inner every->Data points>To its nearest->The Euclidean distance of the nearest neighbor data point, denoted as +.>Distance sets of data points, wherein Euclidean distances in the distance sets of all data points are sequenced in ascending order to form a set +.>The method comprises the steps of carrying out a first treatment on the surface of the Will collect->The element in the cluster is subjected to difference between adjacent element values from left to right, the maximum value of the difference result is obtained, and the Euclidean distance corresponding to the maximum value is used as the clustering feature space region +.>Is>。
The purpose of the above operations is: if abnormal data points appear in a certain clustering feature space region and have great influence on the order information data abnormality of an enterprise, the clustering algorithm of the clustering feature space region needs to keep sensitivity to outliers at the moment, namely a smaller neighborhood judgment radius is set at the moment; for the spatial region with less reaction influence, the clustering algorithm at the moment can properly keep insensitive force on outliers, namely enlarge the neighborhood radius, and classify some data points which are discrete in principle but not very high in degree as core points or boundary points in clusters.
So far, the neighborhood radius in the optimal clustering parameters of any region of the clustering feature space is obtained.
3. And obtaining the minimum number of data points in the neighborhood radius region in the optimal clustering parameters of any region of the clustering feature space.
Specifically, setting optimal clustering parameters for each region in a clustering feature space; the set parameters of the minimum data point number in the neighborhood radius area are as follows:
in the method, in the process of the invention,the +.>The minimum number of data points in the neighborhood radius area in each area; />Representing the +.>A plurality of regions; />Representing the dimension of the cluster feature space, in this embodiment +.>。
So far, the neighborhood radius in the optimal clustering parameters of any region of the clustering feature space and the minimum data point number in the neighborhood radius region are obtained.
And a data detection module: and clustering the order information data of the enterprise according to the optimal clustering parameters to obtain abnormal data.
According to the above, the arbitrary region of the clustering feature space is obtainedOptimal clustering parameters->Andthe method comprises the steps of +.>Clustering operation, and finallyCounting outliers obtained by clustering, and carrying out real-time supervision on abnormal orders in order information data of enterprises corresponding to the outliers; but for->The core points and the boundary points obtained by the clustering algorithm are identified as non-abnormal data points and are not processed; thereby realizing accurate personalized order anomaly supervision and management.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

Claims (6)

1. An order anomaly supervision system based on enterprise management is characterized by comprising the following modules:
and a data acquisition module: acquiring order information data of enterprises;
and (3) constructing a feature space module: analyzing each index data in the order information data of the enterprise to construct a clustering feature space;
the optimal parameter obtaining module is used for obtaining: obtaining a dividing node value according to each index data in the order information data; the method comprises the steps of adaptively obtaining a partition area according to a partition node numerical value pair clustering feature space structure, and obtaining a plurality of clustering feature space areas and serial numbers of each clustering feature space area; selecting different adjacent data points according to the sequence numbers of the clustering feature space regions, and obtaining the neighborhood radius in the optimal clustering parameters of any clustering feature space region according to the distance between the adjacent data points; obtaining the minimum data point number in the neighborhood radius region in the optimal clustering parameter of any clustering feature space region according to the sequence number of the clustering feature space region;
and a data detection module: and clustering the order information data of the enterprise according to the neighborhood radius in the optimal clustering parameters of the clustering feature space region and the minimum data point number in the neighborhood radius region to obtain abnormal data, thereby realizing abnormal supervision and management of the order of the enterprise.
2. The system for supervising abnormal orders based on enterprise management as claimed in claim 1, wherein the step of analyzing each index data in the order information data of the enterprise to construct a clustering feature space comprises the following specific steps:
normalizing the index data return rate to serve as a vertical axis in the clustering feature space; taking the ratio of the total amount of the index data order to the number of the order as a horizontal axis in the clustering feature space; the vertical axis and the horizontal axis form a cluster feature space.
3. The system for supervising abnormal orders based on enterprise management as set forth in claim 1, wherein the obtaining the dividing node value according to each index data in the order information data comprises the following specific formulas:
in the method, in the process of the invention,respectively representing the values of dividing nodes along the horizontal and vertical axes in the clustering feature space, < >>Representing the number of all order numbers of order information data of enterprises; />The +.f in order information data representing the enterprise>Characteristic index data return rate corresponding to the order numbers; />The +.f in order information data representing the enterprise>The number of the characteristic index data orders corresponding to the order numbers; />The +.f in order information data representing the enterprise>The characteristic index data corresponding to the order numbers are the order transaction amount; />Is a linear normalization function.
4. The system for supervising abnormal orders based on enterprise management as set forth in claim 1, wherein the plurality of cluster feature space areas and the sequence number of each cluster feature space area are obtained by the following method:
setting upSequence number of spatial region for clustering feature +.>And (3) taking the value: />The method comprises the steps of carrying out a first treatment on the surface of the Set->The time represents the upper right region in the clustering feature space; set->Representing a region at the lower right part in the clustering feature space; set->The upper left region of the cluster feature space is represented; set->Representing a region at the left lower part in the clustering feature space; thereby obtaining the serial numbers of the four clustering feature space areas and the clustering feature space areas.
5. The system for supervising abnormal orders based on enterprise management according to claim 1, wherein the selecting different adjacent data points according to the serial numbers of the clustering feature space regions, and obtaining the neighborhood radius in the optimal clustering parameters of any clustering feature space region according to the distance between the adjacent data points comprises the following specific steps:
presetting a parameterFor any region in the cluster feature space +.>Acquiring the clustering feature space region +.>Inner eachData points>To its nearest->The Euclidean distance of the nearest neighbor data point, denoted as +.>Distance sets of data points, wherein Euclidean distances in the distance sets of all data points are sequenced in ascending order to form a set +.>The method comprises the steps of carrying out a first treatment on the surface of the Will collect->The element in the cluster is subjected to difference between adjacent element values from left to right, the maximum value of the difference result is obtained, and the Euclidean distance corresponding to the maximum value is used as the clustering feature space region +.>Is>。
6. The system for supervising abnormal orders based on enterprise management according to claim 1, wherein the step of obtaining the minimum number of data points in the neighborhood radius region in the optimal clustering parameter of any cluster feature space region according to the sequence number of the cluster feature space region comprises the following specific steps:
the first in the cluster feature spaceThe calculation expression of the minimum data point number in the neighborhood radius area in each area is as follows:
in the method, in the process of the invention,the +.>The minimum number of data points in the neighborhood radius area in each area; />Representing the +.>A plurality of regions; />Representing the dimension of the cluster feature space.
CN202310921273.9A 2023-07-26 2023-07-26 Order abnormity supervision system based on enterprise management Pending CN116739645A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310921273.9A CN116739645A (en) 2023-07-26 2023-07-26 Order abnormity supervision system based on enterprise management

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310921273.9A CN116739645A (en) 2023-07-26 2023-07-26 Order abnormity supervision system based on enterprise management

Publications (1)

Publication Number Publication Date
CN116739645A true CN116739645A (en) 2023-09-12

Family

ID=87902875

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310921273.9A Pending CN116739645A (en) 2023-07-26 2023-07-26 Order abnormity supervision system based on enterprise management

Country Status (1)

Country Link
CN (1) CN116739645A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116957751A (en) * 2023-09-20 2023-10-27 淄博海草软件服务有限公司 Order service abnormity monitoring method and system
CN117370823A (en) * 2023-12-05 2024-01-09 恒健达(辽宁)医学科技有限公司 Spraying control method and system for agricultural planting
CN117541200A (en) * 2024-01-10 2024-02-09 福建亿安智能技术股份有限公司 Project management method and system based on LTC flow
CN117575546A (en) * 2024-01-17 2024-02-20 北京白龙马云行科技有限公司 Background management system for network vehicle-restraining platform

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116957751A (en) * 2023-09-20 2023-10-27 淄博海草软件服务有限公司 Order service abnormity monitoring method and system
CN116957751B (en) * 2023-09-20 2023-12-19 淄博海草软件服务有限公司 Order service abnormity monitoring method and system
CN117370823A (en) * 2023-12-05 2024-01-09 恒健达(辽宁)医学科技有限公司 Spraying control method and system for agricultural planting
CN117370823B (en) * 2023-12-05 2024-02-20 恒健达(辽宁)医学科技有限公司 Spraying control method and system for agricultural planting
CN117541200A (en) * 2024-01-10 2024-02-09 福建亿安智能技术股份有限公司 Project management method and system based on LTC flow
CN117541200B (en) * 2024-01-10 2024-03-29 福建亿安智能技术股份有限公司 Project management method and system based on LTC flow
CN117575546A (en) * 2024-01-17 2024-02-20 北京白龙马云行科技有限公司 Background management system for network vehicle-restraining platform
CN117575546B (en) * 2024-01-17 2024-04-05 北京白龙马云行科技有限公司 Background management system for network vehicle-restraining platform

Similar Documents

Publication Publication Date Title
CN116739645A (en) Order abnormity supervision system based on enterprise management
WO2021052031A1 (en) Statistical interquartile range-based commodity inventory risk early warning method and system, and computer readable storage medium
CN108564286B (en) Artificial intelligent financial wind-control credit assessment method and system based on big data credit investigation
CN109657547A (en) A kind of abnormal trajectory analysis method based on associated model
CN113572625B (en) Fault early warning method, early warning device, equipment and computer medium
CN113344133B (en) Method and system for detecting abnormal fluctuation of time sequence behaviors
CN109947815B (en) Power theft identification method based on outlier algorithm
WO2019200739A1 (en) Data fraud identification method, apparatus, computer device, and storage medium
CN116562746B (en) Bedding and clothing data monitoring method and related device of bedding and clothing logistics site
CN115358155A (en) Power big data abnormity early warning method, device, equipment and readable storage medium
CN116914917A (en) Big data-based monitoring and management system for operation state of power distribution cabinet
CN112978128A (en) Cold-chain logistics transportation commodity quality monitoring and management system based on big data and image analysis technology
CN117422938B (en) Dam slope concrete structure anomaly analysis method based on three-dimensional analysis platform
CN109976986B (en) Abnormal equipment detection method and device
CN107274025B (en) System and method for realizing intelligent identification and management of power consumption mode
CN117171157A (en) Clearing data acquisition and cleaning method based on data analysis
CN116823125A (en) White spirit storage abnormality early warning system based on multisource data
CN113835947A (en) Method and system for determining abnormality reason based on abnormality identification result
CN113705920B (en) Method for generating water data sample set for thermal power plant and terminal equipment
Jianyao et al. Technical methods of national security supervision: grain storage security as an example
CN111612302A (en) Group-level data management method and equipment
CN114492877B (en) Operation and maintenance analysis method and device of business system
CN117953252B (en) Automatic acquisition method and system for highway asset data
Cheetham et al. Applications of soft CBR at General Electric
Kheirallah et al. Anomaly Detection in a Production Line: an Online/Offline Statistical Learning Approach

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination